This study introduces the Tempotron, a powerful classifier based on a third-generation neural network model, for pulse shape discrimination. By eliminating the need for manual feature extraction, the Tempotron model can process pulse signals directly, generating discrimination results based on learned prior knowledge. The study performed experiments using GPU acceleration, resulting in over a 500 times speedup compared to the CPU-based model, and investigated the impact of noise augmentation on the Tempotron's performance. Experimental results showed that the Tempotron is a potent classifier capable of achieving high discrimination accuracy. Furthermore, analyzing the neural activity of Tempotron during training shed light on its learning characteristics and aided in selecting the Tempotron's hyperparameters. The dataset used in this study and the source code of the GPU-based Tempotron are publicly available on GitHub at https://github.com/HaoranLiu507/TempotronGPU.